[PATCH] D70874: [X86] Add initialization of MXCSR in llvm-exegesis

[Craig Topper via Phabricator via llvm-commits]

craig.topper added a comment.

I'm not sure I understand why this needs to be initialized. Why don't we need to do it for FPCW?



================
Comment at: llvm/tools/llvm-exegesis/lib/X86/Target.cpp:506
+  add(allocateStackSpace(4));
+  add(fillStackSpace(X86::MOV32mi, 0, 0x1f8)); // Mask all FP exceptions
+  add(MCInstBuilder(HasAVX ? X86::VLDMXCSR : X86::LDMXCSR)
----------------
Isn't the default value of MXCSR 0x1f80 not 0x1f8?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D70874/new/

https://reviews.llvm.org/D70874